Estimating Program Impacts on Student Achievement Using "Short" Interrupted Time Series


This paper explores the use of interrupted time-series analysis for estimating the impacts of school restructuring programs designed to increase student achievement. The paper first illustrates the approach and considers its strengths and weaknesses. It then describes how to estimate program impacts and their standard errors from a simple regression model. Next it focuses on the statistical precision of these impact estimates (their minimum detectable effect size) and the research design considerations that affect this precision. The paper concludes by briefly outlining several important issues related to implementing the approach that will be addressed in future research.